This study established the feasibility of using information measurement techniques to analyze reported maintenance data. Failure data on two systems were analyzed: (1) the ASB-4 Bomb/Navigation System of the B-52, and (2) the MA-1 Fire Control System of the F-106. The ASB-4 data (flight-line reports) and the ship data for the MA-1 contained written descriptions of the malfunction symptoms. To make these data amenable to multi-variate uncertainty analysis, a coding scheme designed to retain the grammatical form and content of these symptom reports was used. The F-106 organizational (flight-line) data are already coded (according to ADC 66-28 symptom codes) which make them directly amenable to information analysis. In analyzing the ASB-4 data, it was found that the subject of the reported symptom was most indicative of the unit containing the malfunction, while the predicate phrases and modifiers showed a relationship with repair time. The F-106 data indicated a strong relation between pilot and symptom, although the nature of this relationship cannot be determined from the available data. Some pilots may not be reporting symptoms when they should be. Introducing a training program on Symptom Reporting Procedures should markedly increase systems effectiveness. It appears that a change in job assignment and personnel evaluation procedures could have a large impact on maintenance workload. Additional data are required before specific recommendations can be made, but the results of this study are highly suggestive of the type of data needed to answer the questions raised.